SYSTEM, METHOD, AND PROGRAM FOR SPEECH RECOGNITION, AND SYSTEM, METHOD, AND PROGRAM FOR SPEECH SYNTHESIS

摘要

PROBLEM TO BE SOLVED: To maintain a high recognition rate for speech recognition even with low sound volume without being influenced by ambient noise. SOLUTION: A speech recognition system has a sound information processing part 10 which obtains a sound signal and calculates a sound information parameter according to variation of the obtained sound signal, an electromyographic signal processing part 13 which obtains potential variation on the surface of an object as an electromyographic signal and calculates an electromyographic parameter according to variation of the obtained electromyographic signal, an image information processing part 16 which obtains a photographed picture of the object as image information and calculates an image information parameter according to variation of the object in the picture, a speech recognizing means 20 which recognizes a speech according to the sound information parameter, electromyographic signal parameter, and image information parameter, and a recognition result presenting means 21 which presents the recognition result of the speech recognizing means. COPYRIGHT: (C)2003,JPO